MP Rule 1: Don't Fight A Coil Break -- 'Go-With' It

[Dogpile 10]

No matter what your bias, no matter what the pattern you see, no matter what your indicators say --- when a market has formed a tight equilibrium at the VWAP level: once a market has broken away from this level, keep your trades in the direction of the coil break.

 

http://bp2.blogger.com/_5h-SWVGx6Ms/Rq-gouvXa_I/AAAAAAAAAWk/0MYV9xAg1vw/s1600-h/VWAP+Coil+Break.bmp

[Soultrader 12]

Interesting, what classifies as a breakout from the VWAP level? Is it a break out of the bands? (btw, what bands are those?) Or perhaps I havent been following enough of the strategy threads? Thanks

[Dogpile 10]

The bands are there only to help visualize the coil - not necessarily to define the breakout. I cannot really state a universal definition for when the equilibrium level has been officially broken -- it is up to the trader to decide this. I just wanted to get a thread going to help launch this concept.

 

The point is really that when the market reaches a well-defined equilibrium point, you should recognize this and learn to kind of re-set your thinking that you had before the equilibrium point was established. That is, to get ready to 'go-with' the breakout as there will very likely be at least a short-term sustained directional 'auction' once the market exits its state of equilibrium. At the very least, you should try not to fight this early trend that develops after an equilibrium point has been established.

[jperl 11]

No matter what your bias, no matter what the pattern you see, no matter what your indicators say --- when a market has formed a tight equilibrium at the VWAP level: once a market has broken away from this level, keep your trades in the direction of the coil break.

 

Sorry to have to disagree with you on this Dogpile, but there is more to it than just a break away from the VWAP, tight coil or otherwise. You would be correct if the skew was in the same direction as the breakout, that is < 0 for a short or > 0 for a long. Under any other conditions (skew =0 or breakout in the opposite direction to the skew) you could be in deep doo doo. I've have yet to discuss either of these two other situations, but they will be discussed in future threads of "Trading with Market Statistics"

[nickm001 10]

Dogpile, breakouts in general, are very difficult to trade. If you backtest them, you will find that they fail 60-70% of the time. Money can still be made, but you have to make sure that you lose a little when it doesn't work and stay with the trade for long shot when it does.

[Dogpile 10]

<< there is more to it than just a break away from the VWAP>>

 

Yes, there is ‘more to it’ --- which is really about how good your short-term trading (entry/exit) skills are, not about skew. Unlike most of my set-ups, this set-up is really about the lack of a directional bias for the set-up. As Linda Rashke says, you are better off without a directional bias on the exit of a coil.

 

The pattern here is narrow-range breakout entry – an excellent core technical concept that acknowledges a markets tendency to alternate between ‘trend’ and ‘range-trading’ and specifically, the consolidation that often occurs just prior to a trending move.

 

Trade location is advantageous because this set-up is all about being early-on in a potential trending situation. Proper entry should be somewhere not too far from the equilibrium point with risk defined as somewhere on the opposite side of the equilibrium point (after entry). With these as guideline rules, you simply cannot get into deep doo doo.

[Dogpile 10]

<< If you backtest them, you will find that they fail 60-70% of the time. >>

 

In many respects, a failed breakout is an even better set-up than a coil break since your 'location' will often be better and you may catch a very rapid rejection which puts you immediately into a profit position. The nice thing here is that you monitor the breakout early-on and either it goes or it doesn't. If playing for a failed breakout, then you can watch for failure as price attempts to breakaway from the equilibrium point and catch some of those times the coil does test one way and 'drive' the opposite way. Very similar to 'open-test-drive' concept from Daltons Market Profile books.

 

Yes, good entry/exit trading skills are needed here.

[Dogpile 10]

good example of this concept again today.

 

VWAP std dev bands compress creating conditions for short-term trending auction. look how there is a final test of the VWAP after a push down which gives traders excellent reward-risk location.

 

todays chart also shows how you can make good money in a coil break that eventually fails. a coil-break is a 'go-with' in the short-term. the failure of a coil-break can lead to a bigger move in the opposite direction (this is a Rashke concept -- not mine).

 

todays chart was not quite as clean as recent days in that price did not tighten exactly on the VWAP number -- only near it... but the VWAP number was still a very valid pivot.

 

the Dalton concept here is that there is usually a short-term opporunity to 'go-with' a break away from a balance. I traded a position scalping out 1/2 and putting it back on during pullbacks that didn't break the downtrend -- I also kept a core short that I eventually covered when the market petered out. just an awesome day to trade YM.

 

http://bp3.blogger.com/_5h-SWVGx6Ms/RrD0BuvXbCI/AAAAAAAAAXA/2j-UbbdmT14/s1600-h/VWAP+Coil+Break2.bmp

[Dogpile 10]

I am going to treat this as a blog space for coil-watching for a bit... Here is todays chart:

 

http://bp0.blogger.com/_5h-SWVGx6Ms/RrI82evXbDI/AAAAAAAAAXI/VFvtkAlmTC8/s1600-h/ES+Coil+Break3.bmp

 

Note the differences in the 2 drawn rectangles. The second one is the one to remember to look for -- you have a triangle type of pattern at that point and you have low 5-min ADX -- the market has lost volatility and coiling for a potential trending move.

 

Today was a lower volume NR-7 day so not the best trading environment. I did try a short near point C on the chart and stopped out for a small loss. Note the magenta circle signals the failure of the a-b-c corrective pattern. There was a test down out of the abc up pattern that led to a 'higher low'.

 

Note that this was a 'bear trap' below the VWAP level -- similar to yesterday -- failure to trend in one direction after a coil break gives fuel (short-covering) for a move in the other direction. I did go long above the VWAP level and caught the quick move up. Thus, the small loss enabled me to catch a downside move, if one developed -- but the failure of that pattern enabled me to catch a good move up.

[Dogpile 10]

Price action relative to VWAP is an extremely important concept in trading, in my opinion.

 

As volatility contracts (bands compress) - successful trading strategies will be very different than when volatility is expanding (narrow bands getting wider).

 

Also, I have added an indicator that tracks NYSE volume (not futures volume) versus the previous day. You can see we had a hard early push down today but volume was much weaker than previous day. This was indicator that trending action lacked conviction. This set up compressing std dev bands -- which in turn set the stage for new trending action as the market sought out a new equilibrium level from the 1467.00 area. Note that VWAP was already down versus previous days closign VWAP -- indicating a 'heavy tape'. The market repeatedly failed to break decisively through the 1473.00-74.00 area that market most of yesterdays trading...

[darthtrader 10]

ohh...jerry and dogpile going at it with vwap, this could be the best thread ever.

 

dogpile, any chance you could do some videos like jerry has done? Simple stuff starting out? I would kill to know how you trade. I mean even like a "coil" makes sense seeing it on a chart in retrospect but in real time would be so much more educational.

[Dogpile 10]

jerry got me going on VWAP and I have to say -- it was a missing ingredient in my trading. I have replaced POC-watching with VWAP-watching. The way the market responds to the VWAP price and forms patterns around it is fascinating.

 

I just don't get the PVP part -- I mean I get it in principle but I just don't think its superior to what I was already doing -- but I do find it fascinating that many of the spots I am looking to trade line up with what Jerry is doing statistically. And I want to watch how I might be able to incorporate the PVP-VWAP-StdDev relationship into my trading.

 

To me, pattern-recognition trading isn't optimal by itself -- and statistical trading isn't optimal by itself -- but put the two together and it is quite powerful.

 

I will try a video out at some point and see what kind of response I get.... My view on trading is that us private traders need to work together to take on the massive program trading houses of the world. There are no secrets --- just good trading concepts. Every situation is a little bit different but having core concepts hard-wired into your brain lets you deal with whatever the market throws at you.

 

I know I have improved my trading skills since joining this site. Collaborating is something I am definitely interested in. I will think about how to do something in a video.

[Dogpile 10]

took a shot at a video. this helps explain where I am coming from on this volatility (std dev) stuff. I learned this principle years ago when I was swing trading stocks (my pre-futures days). but the concept is valid for any financial instrument that is liquid and active (has volatility and decent range of movement).

 

quality of video is lacking for some reason. it wasn't bad until I transferred it to youtube. but you can still see what I am doing, I think. I use an AAPL daily stock chart and then bring that back to show what I am talking about with the futures market (S&P futures).

 

to be clear, I am a student of this principle. I haven't seen it presented specifically like this but this is an 'old-school' concept that I think is quite relevant to recent discussions regarding volatility.

 

 

comments/discussion are appreciated...

[Dogpile 10]

a good clean coil did not develop today. nonetheless, following the 'rules/concept' of this idea -- you should clearly not have been shorting this afternoon. The play was to be long or sit out. The market appeared to have substantial resistance above this afternoon but it powered right through it. Had you been following the std dev bands, you would have seen that the bands narrowed and then re-expanded -- the market on this intraday chart was bullish. the issue was whether the market would find sell pressure as it pressed up into the congestion of the last few days.

 

note I have also added an indicator at the bottom (labeled 'VWAP Std DevTrend') that attempts to show when momentum may be forming away from the VWAP level (expanding bands indicating the potential beginning of a trend) -- green for bullish, red for bearish. I am still tweaking this but you can see how it remained green or neutral all afternoon.

[darthtrader 10]

I don't know much about stats beyond trying to read "statistics without tears" but i assume PVP means something because its the mode of the volume distribution?

[darthtrader 10]

to be clear, I am a student of this principle. I haven't seen it presented specifically like this but this is an 'old-school' concept that I think is quite relevant to recent discussions regarding volatility.

 

 

comments/discussion are appreciated...

 

niiiiiice...i have to watch that a few more times before i get it in order to comment much. one thing though is don't you think std dev bands on vwap make more sense as far as market volatility than BBs? I've always been fascinated by BBs, I have Bollinger's book right in from of me. Maybe its just philosphical but it would seem to me the dev bands make more sense when you add volume over just what Bollinger tried to do.

 

some serious MP Dalton idea videos would rule

[Dogpile 10]

<<one thing though is don't you think std dev bands on vwap make more sense as far as market volatility than BBs? I've always been fascinated by BBs, I have Bollinger's book right in from of me. >>

 

the problem with bollinger bands is that they make the assumption that the moving average is the point of 'value' -- which seems kind of flawed.

 

<<Maybe its just philosphical but it would seem to me the dev bands make more sense when you add volume over just what Bollinger tried to do. >>

 

I think the difference in volume-weighting the prices versus just using the closing prices is not such a significant difference.

 

I have grown up on the key concept that 'momentum precedes price' -- such that just as a ball that is thrown up in the air deccelerates before it reverses its flight, a similar tendency occurs in the markets -- a strong directional move will often get a second push in the same direction (which may then be of lesser force). you can measure such with various oscillators. these oscillators are never volume-weighted -- volume is kind of just an adjacent concept. this is how I think of volatility bands -- yes, if the band widens on higher volume that is a more valid move if it doesn't. but a strong momentum push needs to still be respected to some degree. it gets tricky because what if the move is on less than average volume but that volume reading is higher than the immediately preceeding move -- is it valid or invalid then?

 

Many situations have kind of a 'grey area' aspect to them -- I think you have to just develop a sense for the various factors going on and weigh the ongoing buying and selling 'pressures' and not get caught up too much in all of the technical issues.

[Guest cooter]

I have grown up on the key concept that 'momentum precedes price' -- such that just as a ball that is thrown up in the air deccelerates before it reverses its flight, a similar tendency occurs in the markets -- a strong directional move will often get a second push in the same direction (which may then be of lesser force).

 

So is volume the motive force causing all of this? And is the use of MP a means of price-volume discovery?

[Dogpile 10]

<<So is volume the motive force causing all of this?>>

 

not necessarily, IMO... but volume is often important. earlier this year we had these days that just grinded up on low volume and then the market would launch further up and only THEN would the volume come in. volume would come in at bad prices creating a subsequent hard flush the other way ~2 days later. but following volume only left you not believing in the move.

 

 

<<And is the use of MP a means of price-volume discovery?>>

 

MP is good for its concepts. MP concepts are something that helps you with your 'trade location' -- Dalton says "the most important skill you can master to become a successful trader is to distinguish 'price' from 'value'" (pg 100). I don't know if this is the MOST important skill, but its an important one. My results got much better when I began combining my set-ups with concepts of 'value' (trade location). examples of such might be using 'single prints' and buying/selling tails to help pinpoint support/resistance. but momentum is also important. thus something like finding that 'first pullback following strong short-term momentum' combines both concepts -- the first pullback will generally have pretty good location relative to recent 'value' (equilibrium) and you are entering with high reward potential as price has just begun to 'auction' with momentum away from the last equilibrium level -- as it seeks a new level.

[Dogpile 10]

as I mentioned in the video I did over the weekend... the best coil is when volatility (std dev bands) contract AND you can draw a triangle... this occured today around 11:38am EST... the breakout occurs when price moves enough to get the bands to expand.

 

volume was weak on the move relative to previous day -- but you had a non-statistical long bias on days the fed is to announce a decision -- market tends to be safe until an hour or two before the fed announcement -- so the lack of volume was less concerning today.

[darthtrader 10]

ok dog, i soppose pick this up from my PM.

My biggest problem with BBs is they are a std dev measure from a moving average. I'm actually trying to get rid of all thought of ma's from my thinking. I just dont see how they matter. I just think BBs could be done better. Like std dev bands around X atrs. The first "system" i started trading was a linear regression/BB cross that i stumbled on. I can't even tell you what the regression vars were there though.

That was an interesting video with the volume stuff, I would love though as I said though if you could go through some Dalton stuff on a video. I kind of get this stuff, Dalton beyond value pivs make no sense to me.

Have you ever explored Marc Fishers ideas? I have Crabel's ORB book as a pdf, just say the word and i'll find a way to get it to you.